DESCRIPTION: (Scanned from the applicant's abstract) The coactivator binding site of the estrogen receptor (ER) must be generated by hormone binding to the ER, folding the C-terminal helix 12 over the ligand pocket and forming the interaction surface for the LXXLL motif of coactivator proteins, for estrogen hormone signals to properly activate responsive genes in target tissues. Antihormones block the action of estrogens by binding the ER LBD and misfolding helix 12 into a position that occludes coactivator binding. Similarly, the presence and abundance of specific coregulator gene products are also proposed to explain tissue-specific ligand differences, e.g. tamoxifen antagonism in breast and partial agonism in endometrium. Differential activity of estrogenic ligands, including selective estrogen receptor modulators (SEAMs), probably derive from varied cofactor interaction. Single amino acid changes in ER, some with significant phenotypic changes, currently remain under-evaluated. We have developed an assay in yeast to report normal, hormone-induced folding of the ER AF-2 that is very sensitive to mutationa' disruption. Using this assay, ER LBDs have been recovered from tissues and tumors that indeed show altered protein folding. In preliminary data from tumors so far analyzed, 3 have demonstrated changes in ER affecting the p160 coactivator binding site and tamoxifen folded structure. This proposal hypothesizes that significant alterations in estrogen stimulated transcription result from these naturally occurring changes in the estrogen receptor, its co-regulator proteins, or the interaction between them. This has important implications with respect to estrogen and antiestrogen functions in women, particularly those with cancers in estrogen responsive tissues. The project aims are: 1) to determine mechanistic properties of the recovered estrogen receptors, using transcription assays in yeast and mammalian cells; 2) to test cofactor binding to recovered ER alleles with purified GST fusion proteins and in vitro transcription and translation experiments; 3) to isolate via RT-PCR and assay transcription with ER interacting! p160 coactivator clones SRC1, GRIP1, and NCOR cDNAs recovered from tissue samples, and 4) to isolate and identify further naturally occurring ER mutations in vivo, using assays sensitive to ER protein conformation of the AF-2 coactivator binding domain.